ES2400423T3 - Procedure and an automatic landing / take-off control system of a drone on or from a circular landing gate of a specially naval platform - Google Patents

Abstract

Control procedure for the automatic landing / take-off of a drone (4) on or from a circular landing grid (5) of a naval platform (1), a procedure that comprises a step of acquiring the movements of the grid (5), characterized by the fact that it comprises the following stages: - a stage of calculating the average position of the grid (5), - a stage of calculating the predictions of the position of the grid (5), - a stage of calculating the minimums of speed of movement of the grate (5), and - a stage of acquisition of the position of the drone (4) to:. If the drone (4) cannot follow the movements of the grid (5) and if the movements of the grid (5) are small, that is to say less than its radius, apply a localization strategy by tracking the middle position of the grate, while if the movements of the grate are wide, that is, greater than the radius of the grate, apply a location strategy by positioning at the minimum speed of the grate; Y . If the drone (4) can follow the movements of the grate and if the movements of the grate (5) are small, that is, less than the radius of the grate, apply a localization strategy according to the average position of the grate and if the movements of The grate are wide, that is to say greater than the radius of the grate, apply a localization strategy by tracking the position of the grate predicted at the time of landing.

Description

Procedure and an automatic landing / take-off control system of a drone on or from a circular landing gate of a specially naval platform

 [0001] The present invention relates to a method and to an automatic landing / take-off control system of a drone in or from a circular landing gate of a specially naval platform.

[0002] It is known that the problem of landing / take-off control of a drone on a specially naval platform has been raised for several years now.

[0003] In particular, this control must be guaranteed, for example, with sea moving on a small-sized naval platform of the corvette type, for example, whatever the size of the drone, which can also be small and whose movements are then high frequency

[0004] Automatic control processes of this type have already been proposed in the state of the art, for example using laser, GPS, optical, or other means.

[0005] These different means then allow to activate a drone landing according to location strategies that also vary according to different propositions in the state of the art.

[0006] Thus, for example, an already proposed location strategy consists in permanently making the drone position depend on the platform bridge.

[0007] Other location strategies consist in predicting a particular position of the bridge, such as a wave crest, to activate the landing.

[0008] Other strategies also consist of activating the location at the minimum travel speed of the bridge. US2005 / 033489 is an example of a known landing procedure.

[0009] However, none of the solutions proposed so far has given full satisfaction in particular with strong seas.

[0010] The aim of the invention is therefore to solve these problems.

[0011] For this purpose, the invention aims at an automatic landing / take-off control procedure of a drone on or from a circular landing gate of a naval platform, comprising the following steps:

-

 a stage of acquiring the movements of the fence,

-

 a step of calculating the average position of the fence,

-

 a stage of calculating grid position predictions,

-

 a step of calculating the minimum speed of movement of the fence,

-

 a stage of acquiring the drone position to:

•

 if the drone cannot follow the movements of the fence and if the movements of the fence are reduced, that is to say less than the radius of the fence, apply a location strategy by monitoring the average position of the fence, while if the movements of the fence is wide, that is to say, greater than the radius of the fence, applying a positioning strategy by positioning at the minimum speed of the fence; Y

•

 if the drone can follow the movements of the fence and if the movements of the fence are reduced, that is, less than the radius of the fence, apply a location strategy according to the average position of the fence and if the movements of the fence are wide , that is to say superior to the radius of the fence, to apply a location tracking strategy of the predicted grid position at the time of landing.

[0012] According to other aspects of the invention, the procedure and the automatic landing / take-off control system of a drone comprises one of the following characteristics:

-

it comprises a stage of control of the dynamic conditions in speed and altitude of the platform and of the drone, a stage of verification that the drone is effectively in the vertical of the grid and a stage of verification that the position of the grid predicted when it has finished its descent, it is effectively located under this drone, to supply the landing order to the drone,

-

it comprises before the landing phase properly called a meeting phase between the drone and the platform at a predetermined geographical point behind the platform, followed by an approach phase in the course of which the approach path is globally oriented according to the average course of travel of the platform to make an approach behind it,

-

comprises a stage of verification of altitude conditions of the platform, before giving the order to take off the drone, and

-

the stage of control of the altitude conditions consists in calculating some predictions of balancing and pitching of the platform and verifying that these predictions of balancing and pitching of the platform, during the time necessary for takeoff, are inside default threshold thresholds.

[0013] According to another aspect, the invention also aims at a system for performing this procedure.

[0014] The invention will be better understood with the aid of the following description determined solely by way of example and made with reference to the accompanying drawings in which:

-

Figures 1 and 2 represent side and plan views of a naval platform and the approach path of a drone,

-

Figure 3 illustrates the landing of this drone,

-

Figure 4 illustrates a state diagram of an automatic landing procedure according to the invention,

-

Figures 5 and 6 illustrate simulations of landing impacts obtained by performing a control method according to the invention, and

-

 Figure 7 illustrates a state diagram of an automatic take-off method according to the invention.

[0015] A procedure for controlling the automatic landing of a drone on a circular landing gate of a naval platform has been illustrated in Figures 1, 2 and 3.

[0016] This platform is for example designated by the general reference 1 in these figures and therefore comprises a landing zone designated by the general reference 2, provided with a grating adapted to receive for example a maintenance harpoon in the drone position. when perched, in a classic way.

[0017] In fact, the control method according to the invention consists in bringing the drone to a nearby area of the platform thanks to the geo-location information of this platform, then automatically piloting the drone in speed, relatively to this platform. , thanks to a high frequency movement sensor of the type, for example, an optical landing sensor, in order to bring it to rest safely on the landing plate of the platform.

[0018] For this purpose, the position and speed measurements of the platform are treated to develop an approach path of the drone, then the platform / drone position separation measures hybridized with the inertial measures of the platform are processed to develop a platform final drone landing path

[0019] The calculated trajectory is similar to that of the landing of a helicopter to establish an established regime of air circulation around the superstructures of the platform and guarantee the stability of the flight, but also so that the visual safety controls are identical to those of the helicopters for the aviation officer.

[0020] Therefore, it is not only a question of making a permanent dependence on the position of the drone with respect to the bridge of the platform, but of positioning the drone in a particular place above the general movement of the bridge and then expecting the conditions of position, speed and altitude are satisfied.

[0021] Nor is it just about predicting a wave crest to land.

[0022] Indeed, the short-term prediction of bridge positions and altitudes only allows verifying that conditions will always meet at the contact, allowing hardening of the drone's landing gear to absorb the corresponding speeds.

[0023] In addition, the relative distance of the two devices, that is to say the platform and the drone, during the critical phase of the location, is monitored by a non-GPS technology, for example optical and therefore always available and reliable.

[0024] In fact, the recovery of the drone is divided into three general phases illustrated in these figures 1, 2 and 3.

[0025] These phases are the meeting, the approach and the location.

[0026] The meeting of the drone with the platform is a phase of positioning the drone at a point such as fixed GPS of the NED geographical reference (for North East Down).

[0027] This point is placed at safety altitude backwards from the platform on the date of the estimated meeting. This point is designated by E1 in Figures 1 and 2.

[0028] The approach is a sequence that allows the drone to enter the bridge in the relative wind direction.

[0029] The information on the average geo-location of the platform allows defining a setpoint approach path for the drone, by aligning the two devices from point E1.

[0030] The approach path is oriented globally according to the middle course of the platform to make an approach behind this platform facing the helicopter hangar of this for example.

[0031] Next, the drone path is finely oriented towards the infinite wind upstream of the platform if it is comprised in the wind chute authorized by the drone / platform. If it is not, the wind contract is not fulfilled by the platform's aviation gateway and the ship's trajectory must change to observe this wind gauge.

[0032] This approach path passes through a point E2 illustrated in these figures 1 and 2, the drone approaching the platform in distance and altitude.

[0033] As the drone continues its approach to the platform, it enters the field of vision of the optical separation measuring means implanted in the platform, this field of vision being designated by the general reference 3 in these figures 1 and 2.

[0034] Then the precise pilot phase of the drone is started with a view to its landing.

[0035] The drone is then brought under the control of the means that form the separation sensor to chain position setpoints W, T1 and T2 with expected waiting phases at each of these points, to verify if the dynamic conditions are met. of the landing and in particular the relative speeds and altitude angles between the bridge and the drone.

[0036] The location strategy performed will then depend on the movements of the platform and the dynamics of the drone.

[0037] Thus, if the drone cannot follow the movements of the fence and if the movements of the fence are reduced, that is to say less than the radius of the fence, then a location strategy will be applied by monitoring the average position of the fence. , while if the movements of the fence are wide, that is to say greater than the radius of the fence, then a positioning strategy will be applied by positioning at the minimum speed of the fence.

[0038] If the drone can follow the movements of the fence and if the movements of the fence are reduced, that is, less than the radius of the fence, a location strategy will be applied according to the average position of the fence and if the movements of The fence is wide, that is, greater than the radius of the fence, a location strategy will be applied by monitoring the position of the fence predicted at the time of landing.

[0039] The vertical descent order is given when the drone is in T2 and the following conditions are satisfied simultaneously:

1) Dynamic conditions in speed and altitude are satisfied, depending essentially on the resistance of the landing gear and the height of the center of gravity of the drone.

2) The drone is measured in the vertical of the grid by the means that form the optical separation sensor.

3) The position of the fence is predicted under the drone when it has finished its vertical descent, that is for example in less than 5 seconds.

[0040] To observe the movements of the grid and determine the location strategy, the principle of extraction of the movements of the grid is used in the average pseudo-inertial reference of the platform and in the short-term prediction of the position of The grid uses classic signal processing techniques to statistically identify the behavior of a physical system using a technique for estimating the coefficients of a stable stable oscillating filter, to predict the position of the grid in a pseudo inertial reference that goes to the platform speed.

[0041] The numerical conditioning of the coefficients of the filter is decisive and taking into account the random components of the movement of the platform, these techniques allow a reliable prediction in a few seconds, which is enough to validate that the landing will be carried out correctly.

[0042] All of these means are already well known in the state of the art and will not be described in more detail in the following.

[0043] As illustrated in Figure 3, this allows the drone, designated by the general reference 4 in this figure, to be carried over the bridge of the platform 1 and in particular above the landing gate 5 of this .

[0044] Once posed, a means of fixing the drone to the bridge can be activated, such as a harpoon on the landing gate.

[0045] This is illustrated, for example, in Figure 4, where it can be seen that the different orders sent to the drone and more especially to its automatic piloting means, order it in 10 to be placed in point E1, in 11 to be placed in point E2, at 12 be placed at point W, where the control of its position passes from the GPS system to the optical separation sensor at 13.

[0046] The drone then moves on in T1 as illustrated in 14 and after a waiting phase in 15, descends in T2 in 16 and after a waiting phase in 17, perches on the bridge in 18, before activate the fixing means such as for example the harpoon in 19.

[0047] Figures 5 and 6 illustrate simulations of trajectories and impacts on the landing, in the form of a Monte Carlo-type simulation sweep, of 50 landings with sea force 5 with a swell of 165 ° (15 ° in the front part).

[0048] These figures show that in 50 landing simulations, 39 have been achieved at the first.

[0049] 11 landings are next to the fence in the first inn and in this case the drone is taken back to take off and try again to land.

[0050] Trials have been initiated in magnitude of nature and have confirmed the reliability of this control procedure.

[0051] Likewise, the method according to the invention also comprises a stage of verification of altitude conditions of the platform before giving the takeoff order to the drone.

[0052] This control stage consists in calculating balancing and pitching predictions of the platform and verifying that these balancing and pitching predictions, during the time necessary for takeoff, are within predetermined threshold thresholds such as It is illustrated in Figure 7, or after a starting stage illustrated in 20, the drone is put on hold at 21 before activating its automatic take-off at 22, from which the drone is considered as being on mission at 23.

[0053] Thus, automatic take-off is conditioned by the altitude conditions of the platform at the time of take-off, to prevent the drone from taking off too steeply and reaching an unexpected horizontal speed.

[0054] The principle of take-off then consists in going to point E1 after the take-off order knowing that the take-off order is only launched after a continuous prediction of the altitudes of the rolling and pitching platform inside the authorized thresholds during the time necessary for takeoff.

[0055] The system for carrying out this procedure then comprises a certain number of data acquisition means such as movements of the grid and calculation such as the average position of this grid or also predictions of the position of the grid and minimum travel speed of this.

[0056] It also includes means of acquiring the position of the drone and means of transmission of control orders to the drone and more especially to the means of automatic piloting of the latter in order to take it to land and / or take off safely.

[0057] To present these means based on inertial data acquisition centers, GPS systems, optics or other classical structures, will not be described here again.

[0058] In fact, these means may have any of the appropriate structures that integrate computer programs for performing the different steps described above.

Claims (6)

1. Automatic landing / take-off control procedure of a drone (4) on or from a circular landing gate (5) of a naval platform (1), a procedure comprising a stage for acquiring the movements of the gate (5 ), characterized by the fact that it comprises the following stages:

-

 a step of calculating the average position of the fence (5),

-

 a stage for calculating grid position predictions (5),

-

 a step of calculating the slider travel speed minima (5), and

-

 a stage of acquiring the position of the drone (4) for:

. if the drone (4) cannot follow the movements of the grid (5) and if the movements of the grid (5) are reduced, that is to say less than the radius of this, apply a location strategy by monitoring the average position of the fence, while if the movements of the fence are wide, that is to say greater than the radius of the fence, apply a positioning strategy by positioning at the minimum speed of the fence; Y . if the drone (4) can follow the movements of the fence and if the movements of the fence (5) are reduced, that is, less than the radius of the fence, apply a location strategy according to the average position of the fence and if movements of the fence are wide, that is to say greater than the radius of the fence, to apply a tracking strategy by tracking the position of the fence predicted at the time of landing.

2.

Automatic landing / take-off control method of a drone according to claim 1, characterized in that it comprises a stage for controlling the dynamic conditions in speed and altitude of the platform (1) and of the drone (4), a verification stage that the drone (4) is effectively in the vertical of the grid (5) and a verification stage that the position of the grid (5) predicted when it has finished its descent, is effectively located under this drone , to supply the landing order to the drone.

3.

Automatic landing / take-off control procedure of a drone according to claim 1 or 2, characterized in that it comprises, before the landing phase itself, a meeting phase between the drone (4) and the platform (1) in a predetermined geographical point behind the platform, followed by an approach phase in the course of which the approach path is oriented globally according to the average course of travel of the platform to make an approach behind it.

Four.

Automatic landing / take-off control procedure of a drone according to any of the preceding claims, characterized in that it comprises a stage of verification of altitude conditions of the platform (1), before giving the take-off order to the drone ( 4).

5.

Automatic landing / take-off control procedure of a drone according to claim 4, characterized in that the stage of control of altitude conditions consists in calculating balancing and pitching predictions of the platform (1) and verifying that these predictions of balancing and pitching of the platform, during the time necessary for the take-off of the drone (4), are within predetermined threshold limits.

6.

Automatic landing / take-off control system of a drone (4) on or from a circular landing gate (5) of a naval platform (1), comprising means of acquiring the movements of the gate (5), characterized by the fact that it comprises means for calculating the average position of the grate (5), means for calculating the predictions of the position of the grate (5), means for calculating the minimum travel speed of the grate (5), means for acquiring the position of the drone and means for verifying altitude conditions of the platform, to carry out the method according to any of the preceding claims.

RELATIVE WIND FOV SENSORS RELATIVE